Spinoff2: Using orientation to control position 1

[fsincox]

The whole idea of using orientation to control location has also come up in my past work as the “GD&T guy”. A very well respected senior inspector proposed the use of parallelism to control the location of shaft keyways. His idea was that you would line up the shaft diameter (the primary datum) and then roll the keyway, itself (the secondary datum) to center and proceed to measure the resulting parallelism of the opposed sides of the keyway. His rational was that the resulting parallelism of the sides of the keyway in this representation of the installed state was important to the key not working its way out under heavy loading. I have to admit I did not feel it was technically illegal, what do you think?

 

[Belanger]

But Evan, on pmarc's drawing datum feature B was just the slot's left wall. Otherwise, A-OK!

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[axym]

J-P,

That's not how I understood it. In pmarc's drawing, the datum feature symbol is directly in line with the size dimension. So I thought that datum feature B was the slot. Those darn drafting-based rules!

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[axym]

pmarc,

The "back-door location" effect relies on the fact that the simulator for B is directly in line with the simulator for A. In other words, simulator B is basically located to simulator A. If this were not so, then the simulator could shift to accommodate the location/orientation error of the slot. The end result would be that datum centerplane B would pass through datum axis A but be parallel to the unrelated AME of the slot. Using this DRF, the mislocated/tilted slot would meet the print.

I am not sure where ISO stands on this, but basic location of simulators has only been required in Y14.5 since 2009. See Section 4.5.2 on page 53:

Datum feature simulators shall have the following requirements:

a) perfect form.
b) basic orientation relative to one another for all of the datum references in a feature control frame.
c) basic location relative to one another for all of the datum references in a feature control frame, unless a translation modifier or movable datum target symbol is specified
d) movable location when the translation modifier or the movable datum target symbol is specified

But it wasn't this way in '94. From what I remember, Y14.5M-1994 didn't make any explicit statements but certain figures indicated that the simulators were not basically located. But Y14.5.1M-1994 spelled it out clearly, that lower-precedence datums were basically oriented but not basically located. See Section 4.3.2 c) and Fig. 4-2 in Y14.5.1M-1994. The geometry Figure 4-2 is very similar to the situation we are discussing in this thread, except that the slot is a tertiary datum feature. But the method of clocking the DRF to the slot is clearly shown, and orienting the slot sides to this DRF would not give back-door location.

So Frank and his inspector are correct, but maybe they've only been correct since 2009. ;^) Again, the rule might be different in ISO and may have thrown you off. Do you know how ISO handles this issue, pmarc?

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[Belanger]

Ugh, I noticed that after my post! You are right; the datum is the center of the slot, not the one wall.
Add my name to the list of flubbers in this thread

The 1994 standard does mention that a datum center plane such as our situation is to be in line with the previous datum. It's in the text of Figs. 4-15 thru 17.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[pmarc]

Evan, J-P,
Thanks for your try, but I do not buy it. I mean, the explanation on why datum centerplane B must be in line with datum axis A does not convince me.

The figures you both are refering to are different from my sketch in one veeeeeeeeeeeeery important aspect: in all of them there is a locational relationship defined (i.e. positional tolerance) between slot's centerplane and higher precedence datum(s). If that was the case in my sketch reflecting Frank's dimensioning scheme I would not have any remarks and would never say to him that his interpretation was wrong.

But since nothing defines a relationship between datum features A and B, I do believe nothing constrains the translation of datum centerplane B to datum axis A.

 

[axym]

J-P,

You're right about the datum plane being in line in '94, my recollection was wrong. Serves me right for relying on memory and not actually looking it up. This was a direct contradiction between the math standard and Y14.5. Major problem. We'll have to make sure to remove things like that in the next version of the math standard, because they seriously affect its credibility and usability.

So Frank and his inspector were correct all along, unless you go with the Y14.5.1 meaning. But who reads the math standard anyway? ;^)

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[axym]

pmarc,

So you're saying that a basic relationship between the datum features is not sufficient, that the relationship must be toleranced or we do not get the basic relationship between the simulators? You might be right, but I'll have to think about that one.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[pmarc]

Yes Evan, this is what I am saying.

 

[Belanger]

I'm not following this latest angle -- isn't it accepted that the simulator for datum B must come down toward the shaft while in line with datum A? I understand that there is no tolerance on the location of the slot, but we are simply trying to simulate the datums in order to check the two FCFs.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[Belanger]

FYI -- if it helps, I'm looking a para. 4.11.4 (d) in the 2009 standard and 4.5.3 (c) in 1994.

It's interesting that the 2009 standard says that the simulator for datum B in our situation "must be oriented and/or located to the primary datum feature's datum feature simulator."

That whole thing about and/or might play into the different interpretation that pmarc is holding to. I for one will try to noodle this thing again.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[DeanD3W]

pmarc,
Evan pointed out Y14.5-2009 section 4.5.2 already... In that section the way to have the effect you describe in your sketch is provided... Per sections 4.5.2 (c) & (d) the translation modifier would need to follow the datum feature B letter in the parallelism tolerance(s) applied to the slot. Without that modifier there is nothing that I know of in the standard that will make the simulator for datum feature B anything but fixed and centered relative to datum A.

Dean

http://www.d3w-engineering.com

 

[pmarc]

Guys,
Please have a look at the attached sketch. In order to show my point as clear as possible, I modified Frank's part a bit; I removed both parallelism callouts and added two holes just to have datum B referenced somewhere on the print.

http://files.engineering.com/getfil...4a-2afe47504df9&file=keyway_parallelism_2.JPG

My questions are following:
- How will the datum feature simulators A & B look like for holes position tolerance check (especially mutual relationship of simulators) depending on the case?
- Will there be any differences amongst the three?

 

[CheckerHater]

pmarc,
Could you loosen up some of the requirements?
The whole point could be lost when trying to control evrything at RFS.

 

[CheckerHater]

I have to post it before sleeping pill wears out :
According to Y14.5-2009 Parallelism can be replaced with Angularyty.
Gentlemen, I am giving to you Case#4 - The smaller the angularity the better the location:

 

[axym]

pmarc,

My first instinct is that the simulators will be an adjustable size sleeve and an adjustable size block, with the block being basically located to the sleeve. There would be no difference for the 3 cases. I do not know of any Y14.5 references that indicate that the simulators would have different mutual constraints depending on their tolerances.

At the same time, Y14.5 does say that the relationship between the datum features shall be controlled. This was only a suggestion in Section 4.3.3 in 1994, but was much more strongly mandated in Section 4.9 in 2009. Case #1 and Case #2 would not meet the criteria specified in 4.9 b), because they do not control the location of the secondary datum feature to the primary.

One could say that Case #1 and Case #2 are incomplete drawings anyway, because the location of the slot relative to the OD is not controlled. But I'm wondering what the interpretation would be if this was controlled by a general Surface Profile tolerance. Is that enough, or does there need to be a location control on B that references A?

Also, does the fact that datum feature B is referenced RMB make any difference? It shouldn't, but if datum feature B was referenced at MMB then the relationship between the datum features would have to be more clearly defined. Otherwise, the MMB and simulator size could not be calculated.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[Belanger]

For Pmarc's scenario:
Case 3 is the easiest -- a datum feature simulator must be in line with axis A as it comes down. This is clear from the standard (Fig. 4-15) and there is a basic relationship between A and B.

I suppose Cases 1 and 2 would be the same simulator for each other, because they both do not have GD&T to locate the slot to A. However, the location is implied to be in line with axis A. So the million-dollar question is: what is the tolerance on that implied side-to-side location? Title block? That could be dicey if the title block separates the tolerances by the number of decimal places.

So perhaps 1 and 2 are unclear drawings (as is the original one that started all this), because the relationship between the datum features is not complete. Here's a similar analogy: In Fig. 4-20 (c) of the standard, what would be the size of the datum feature simulator for A if there were no perpendicularity tolerance given? It wouldn't be 16.0. There has to be some tolerance on the orientation in order to properly describe the simulator for B in Fig. 4-20.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[pmarc]

Evan, J-P,
Of course, cases 1 & 2 are incomplete by intent. I just wanted to know whether in your opinion datum feature simulators A & B were the same for all 3 cases or not. I see that you agree they are the same.

My last try: what about implied basic zero dimension between axis of datum feature A and the centerplane of the slot B? Does it apply here [based on of 1.4(k) of Y14.5-2009] or not?

 

[Belanger]

Pmarc, I kind of see what you're saying. There is an implied zero relationship between A and B, but it's not basic, since no location tolerance is given to necessitate a basic dim.

But that's the point; the drawing is incomplete, and thus our whole dilemma is unanswerable, in my view. See my analogy about Fig. 4-20, if there were no perp tolerance.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[pmarc]

J-P,
And that was also my point regarding Frank's example. Like I said in my post from 5 Feb 12 12:37: "The figures you both are refering to are different from my sketch in one veeeeeeeeeeeeery important aspect: in all of them there is a locational relationship defined (i.e. positional tolerance) between slot's centerplane and higher precedence datum(s). If that was the case in my sketch reflecting Frank's dimensioning scheme I would not have any remarks and would never say to him that his interpretation was wrong."

Since Frank's dimensioning scheme is incomplete (i.e position tolerance is missing) I cannot understand why you, Evan and Dean are saying that datum feature simulator B is fixed in location relative to datum feature simulator A.

And yes, I agree with you when you are refering to 4-20 and a potential lack of perpendicularity callout. Paragraph 4.9 is very specific on that.

 

[DeanD3W]

pmarc,
The presence or absence of tolerances on the datum features has no bearing upon the requirement that datum feature simulators be "perfectly" oriented and located with respect to each other... Unless you apply a sliding modifier per Y14.5-2009, the simulator for the slot, datum feature B, must be aligned with datum A.

Dean

http://www.d3w-engineering.com